SUMMARY (Project I) Invariant pathological features of Alzheimer s disease (AD) are changes in astrocytes, microglia and innate immune signaling factors that collectively denote altered innate immune activation states within the brain. Although there has been past interest in therapeutic strategies targeting inflammation and innate immunity in AD and mild cognitive impairment (MCI), there has been more limited recent activity in this therapeutic area even though targeting innate immunity may provide therapeutic advantages. A major challenge for both therapeutic development and pathogenic understanding is the lack of facile tools to non-invasively evaluate innate immune activation states in humans with varying degrees of AD pathology and cognitive and functional impairments. Recently developed algorithms for analyzing diffusion magnetic resonance imaging (MRI) data suggest that free-water imaging that assays extracellular volume can provide a surrogate measure of inflammation throughout the entire human brain. In this project we will explore in parallel studies of mouse models and humans the hypotheses that free-water imaging 1) is a marker of inflammation in mouse models, 2) will distinguish humans with AD, late-MCI, early-MCI, and pre-MCI from healthy controls, and 3) provides valuable prognostic data regarding cognitive changes over time that are specifically associated with innate immune activation in the brain. Two aims are proposed. Aim 1: Use novel models of chronic CNS neuroinflammation and mouse models that develop AD relevant pathologies (A? and tau) to provide an enhanced understanding of the biological basis for changes in MRI-based detection of free-water and examine how modulation of inflammatory responses and other pathologies in these models alters free-water levels. Aim 2: Test the hypothesis that extracellular free-water levels derived from the diffusion MRI data that will be acquired as part of the Clinical Core B assessments distinguishes AD, late-MCI, early-MCI, and pre-MCI, from controls (CN). These translational studies are well-supported by preliminary data, highly integrated into the overall ADRC and leverage the broad experience of the PI and Co-Is with respect to AD and neuroinflammatory animal models, small animal MR imaging, and human subject MR imaging. By further establishing the pathological basis of MR-based markers purported to track inflammation and evaluating these in a large clinical cohort these studies will inform on the utility of these methods as biomarkers and potential theragnostic markers for AD.